[Pkg-shadow-devel] faillog reset

[Peter Vrabec]

On Tuesday 24 November 2009 06:25:12 pm Nicolas François wrote:
> Hi,
> 
> On Tue, Nov 24, 2009 at 04:36:26PM +0100, pvrabec at redhat.com wrote:
> > I just want to ask if there is a reason why reset() in faillog iterates
> > thru "all" UIDs. Can we skip UIDs that a free? I mean there is no user
> > with such a ID.
> 
> There is a need to reset the faillog entries for removed users.
> 
> But I agree we could differentiate "faillog -r" and "faillog -r -a"
> 
> For reset, it's probably easier / faster not to care and reset everything.
> 
> For setmax, or set_lock there can be some drawbacks as filling the file
> (instead of having a sparse file), so it might be better to have the 2
> behaviors (it can also be useful to lock or setmax for future users).
> 
> 
> Peter, do you think this should be changed?
yeah, I think it would be fine to have two behaviors.

> In that case, -r, -l, -m should be changed. lastlog should probably be
> changed too (it is usually very similar to faillog).
> 
> Best Regards,
>